Menu

The provide performance-enhancing effects. However, studies also

0 Comment

The
rise of ‘new public management’ popularized doctrines of explicit standards and
measures of performance at both the international and national levels.  Performance management became the preferred
approach for public managers to motivate and increase efficiency.  The popularity is evidently shown by the
increase in usage of performance management systems such as international
governance rankings, which doubled every ten years during the 1970s to 2000s
(Hood, 2012).  A number of studies reveal
that performance management can improve efficiency and provide
performance-enhancing effects.  However,
studies also show that efficiency is usually gained at the expense of public
value as a result of unintended consequences caused by implementing performance
measurements.  Thus, there is a tradeoff
between increasing efficiency and creating public value.  This essay will illustrate why ‘public
management techniques such as performance management increases efficiency at
the expense of creating public value.’

 

There
is an overwhelming consensus that the private sector is better managed than the
public sector.  Performance management in
the private sector, specifically the use of financial ratios, plays a
significant role in the success of private companies by providing a clear goal
or objective.  Thus, it is not surprising
that a number of public managers want to replicate the success by learning from
private companies and applying the same methods to the public sector.  The rise of public management by numbers was
mainly driven by claims of its performance-enhancing effects.  According to Barber, performance management
is viewed as the best approach to increase efficiency because it provides clear
objectives, definitions of success, and accountability (Barber, 2015).  However, financial ratios used in the private
sector cannot be applied to the public sector in the same manner, as the main
objective of governments is not to maximize profits but to maximize public
value.  Although there is no consensus on
the definition of public value, Moore and Khagram define it as “the use of
government assets to produce a good and just society” (Moore & Khagram,
2014).  Public management by numbers in
the public sector is not as straightforward as measuring profits in the private
sector, but rather, it must capture “the public’s aspirations and concerns as
well as the procedural norms and values associated with good public sector
governance” (Alford, Douglas, Geuijen, & Hart, 2016).  Given the complexity of measuring public
value, views and discussions in literature on the approaches are mixed.  

 

Due
to its extensive use in the public sector, performance measurement has been
both widely criticized and praised for its effects.  Barber claims that performance measurement is
a powerful tool to motivate members of an organization to increase performance by
providing clear objectives and maintaining accountability of government
officials (Barber, 2015).  There is
evidence that performance measurement is effective in increasing efficiency
within the public sector.  For example, target
setting has led to the reduction of hospital waiting times and better
allocation of public funding (Li, 2015). 
Advocates of performance measurement such as Osborne and Gaelber assert
that without measurement, it is not possible to determine whether the
organization is successful or failing. 
Without recognizing failure, the organization cannot correct its mistakes
and learn from them (Blaug, Horner, & Lekhi, 2006).  Conversely, oppositions to performance
measurement argue that public value is destroyed due to unintended consequences
that arise from implementing such measures. 
Although unintended consequences can be both beneficial and harmful,
performance measurements in public services often lead to gaming responses that
destroy public value, namely the ratchet effect, threshold effect, and output
distortion (Li, 2015).  For instance, when
a target was set for the emergency room waiting time, patients were unfavorably
kept longer in ambulances because the waiting time in the ambulance was not
targetized (Hood, 2012).  Furthermore, Hood
offers a different viewpoint, suggesting that the effects of measurement are dependent
on the context and culture of the organization, namely, hierarchist, fatalist,
egalitarian, and individualist.  In other
words, if the performance measurement method chosen is suitable to the context
and culture of the organization, performance-enhancing effects will occur.  However, it is not always easy to determine the
suitable measurement method, and that “the costs and benefits of performance
measures are seldom, if ever, computed in a way that would allow us to assess
when the performance benefits of such systems exceed the costs and vice versa
(Hood, 2012).”  Although it is evident that
performance management can lead to increases in efficiency, the unintended
consequences that result from measurement often destroy public value, outweighing
the benefits of efficiency gained. 

 

One
of the main reasons performance measurement destroys public value is the
occurrence of unintended consequences.  Implementing
measurements create significant changes within an organization, where there are
a large number of stakeholders and members, making unintended consequences
inevitable.  Thus it is impossible to anticipate
all possible outcomes due to the variation of interactions among members of the
organization.  Unintended consequences
commonly found in public management by numbers are gaming responses, the most
common types being the threshold effect, ratchet effect, and output
distortion.  The threshold effect refers
to people’s tendency to become disincentivized once they have already achieved
the set target.  Whereas the ratchet
effect is the tendency for people to not perform at their optimal level if they
expect future targets to be set incrementally higher, thus making it more
costly and difficult to achieve in the future. 
Lastly, output distortion describes the motive for people to distort
measured outcomes to make it seem more favorable (Hood, 2012).  The type of gaming responses provoked by
members of an organization will depend on the method of measurement used.  In addition to the three common types of gaming
responses, performance measurement can also lead to other unintended
consequences such as selective focus, in which members of the organization only
focus on aspects that are measured while ignoring others that may be crucial to
creating public value.  Performance
measurement promotes the notion “what’s measured is what matters” (Li, 2015).  Different types of measurements will lead to
different unintended results.

 

To
begin with, targets are one of the most common and simplest forms of
performance measurement.  Targets involve
setting a minimum quantitative threshold of performance or activity to be met
by the organization.  This method of
measurement provides a clear definition of success, making it an effective motivating
tool for members of the organization.  Targets
enhance efficiency by enabling the organization to allocate its resources effectively
and to align objectives.  Despite the
ability of targets to increase efficiency, this measurement method exposes the
organization to a number of gaming responses. 
An example of this phenomenon is conveyed in a study conducted on the
use of target regimes in a Chinese local government by Jiayuan Li.  The study was done in Guandong, where
higher-level government officials control local officials and rely on target
systems to evaluate their performances.  In-depth
interviews using probing questions with local officials were conducted to
obtain detailed accounts of responses to processes of performance measurement
in various public services departments, namely public security, environmental
protection, food safety, education, and etc. 
The analysis of the interviews revealed that local officials admit to
distorting performance data that is reported to upper-level government
officials due to pressures to meet targets. 
Output distortion was exposed across departments.  For instance, in order to meet the
requirements set by higher-level officials, the statistical department at the
district level colluded with townships to distort economic data.  The department would knowingly accept
falsified data from townships in order to meet targets.  Although the officials in the statistical
department were fully aware that their main role was to provide authentic data
for government use, the incentive for personal reward and fear of not meeting
targets were stronger to provoke perverse behavior.  Not only does the use of falsified data diminish
the public’s trust and the integrity of the statistical department, but the
false data may also lead to harmful policy decision-making.  Bad policy decision-making can negatively
impact the economy and society in the long run, destroying public value.  In addition to output distortion, the
research also revealed the presence of ratchet effects.  The targets for crime reduction are set based
on the performance of the previous year. 
Hence the prospect of higher future adjustments distorts police behavior
to underperform so that they can exert less effort in the future.  This practice cultivates a culture within the
department where using a ‘conservative strategy’ to report results is the norm.  Society is worse off as police officers are
not motivated to further reduce crime, sacrificing public safety.  Furthermore, a third problem created by target
setting in Chinese local government is selective focus.  The study reveals that police officers only
paid attention to indicators that were measured, and ignored those that were
not.  Given the overemphasis on crime
reduction measurements within the police force, the officers allocated all of
their efforts toward crime reduction at the expense of crime prevention (Li,
2015).  Sacrificing crime prevention is
costly because such preventive measures may be more effective in reducing
crimes in the long run.  Evidently, due
to the distorted incentives created by targets, Chinese local officials behave
in a way that destroys public value. 
Again, the use of targets may increase efficiency in the short-run but
it does so at the expense of public value.

 

The
second type of performance measurement is rankings, which involves comparing
quantitative measurements relative to others. 
Rankings are competent at increasing efficiency by using competition to
incentivize organizations to do better than their rivals or peers.  Rankings are commonly used in the public
sector to evaluate the progress of public services relative to others (Hood,
Dixon, & Beeston, 2008).  Similar to
targets, the use of rankings is susceptible to unintended consequences that impede
the creation of public value.  A
prominent example of rankings used in the public sector is in schools.  School rankings are used as an indicator of
education quality, usually based on the performance of students on standardized
tests.  Rankings are often publicized by
the media, which draws more attention and pressure on schools.  As a result, publishing rankings can lead to
more uncertainty and invoke inappropriate behaviors from school officials
(Karsten, Visscher, Dijkstra, & Veenstra, 2010).  For instance, in 2002 President George W. Bush
introduced mandatory standardized tests as part of the No Child Left Behind Act
(NCLB), with the intention to increase quality of education and accountability
of schools.  The schools of each district
were ranked based on students’ standardized test scores, graduation rates and
teacher ratings.  School rankings were an
effective tool to motivate both faculty and students to perform better as poor
performing schools would be vulnerable to investigation and closure.  Thus, it is not surprising that the implementation
of performance measurement had the largest impact on the worst performing schools,
mainly located in low-income districts.  The
test scores became the main focus in the schools, even to the point where
teachers were teaching the tests to students. 
Schools in Atlanta in particular saw remarkable improvements in their
test scores; the improvements were so extraordinary that people became
suspicious.  Analyses were done on the
Criterion Referenced Competency Tests (CRCT), which was a state-administered
standardized test used in Atlanta, revealing the unlikeliness of achieving such
scores in a short period of time without cheating (Whack, 2015).  This led to an investigation by the Georgia
Bureau of Investigation, who discovered that 44 out of 56 schools cheated on
the tests by having teachers change the answers to the tests so that the
students would pass the exams.  The
cheating scandal became one of the worst the United States have ever uncovered,
involving over 180 employees, including 38 principals (Blinder, 2015).  The emphasis and pressure on the rankings and
standardized testing created a distorted incentive to principals and teachers
that led them to cheat.  This example
clearly portrays the use of measurement creating unintended consequences that cultivated
a cheating culture within schools and exacerbated the learning conditions for
students.  Faculty and teachers were fixated
on test scores that they neglected other aspects of education that will provide
necessary skills for students to be successful in the future.

 

The third type of performance measurement
is intelligence.  The intelligence
approach refers to the use of quantitative data to provide insights and
information.  This approach focuses on
the learning process and the gathering of information, unlike rankings and
targets.  In other words, intelligence is
“information that is collected or pieced together for purposes that cannot be
clearly specified in advance” (Hood, 2012). 
Given that the intelligence approach is not directly linked to rewards
or punishments, the ratchet effect and threshold effect are not commonly found
using this method.  However, similarly to
other methods of performance measurement, intelligence is also vulnerable to
unintended consequences.  For instance, in
2005, the Netherlands introduced a system called the Diagnosis-Related Group
(DRG) as part of a health care reform. 
The DRG system was designed to classify patients into clinically and
cost homogenous groups.  The system was
implemented to increase efficiency and accountability of health care providers,
as well as introduce market processes into the health care sector.  A study was conducted by Kerpershoek,
Groenleer and Bruijn to unveil the unintended responses of using the DRG system.  Both qualitative and quantitative data were
collected through the use of in depth interviews with medical professionals
from a number of hospitals and the examination of documents from the system.  The subjects were asked questions about how
they used the DRG system in their daily routines.  A two-step coding procedure was used to
analyze the interview, labeling the subjects’ answers according to word choice
and placing the associated unintended behaviors into categories.  The analysis revealed several unintended
consequences.  The first was that medical
specialists would register different diagnoses and treatments from the ones
given to the patients into the system.  A
likely explanation for this behavior was that the medical specialists did not
want their patients to pay for treatments that were not covered by insurance,
and the reimbursement amount would be affected by the registered diagnosis and
treatment.  Another unintended behavior
caused by the DRG system was cherry picking, where patients were selected
according to their risk profiles. 
Hospitals would accept patients that had low complication risks and
would defer high-risk patients to other practices.  In addition to the patient’s risk profile,
patients were also selected based on costs; those with high cost diagnostics
were unlikely to be chosen.  Cherry
picking was also done to balance profitable and unprofitable DRGs, as more volume
of the profitable DRGs would be produced (Kerpershoek, Groenleer, & Bruijn,
2014).  The utilization of the DRG system
may enhance efficiency by increasing the ease of access to information for
hospitals and medical service providers, but it also leads to unwanted behavior
such as registering inaccurate information and cherry picking. This destroys
public value because false data may be used to implement policies that create
negative effects to the health care sector. 
Moreover, patients are deterred from receiving the best care.  Practices that are best equipped to treat the
patients may turn them away due to high associated costs accessible in the
system.  Unlike targets and rankings, the
motivation behind each distorted behavior is not easily discernible when using
the intelligence method, making it even more difficult to anticipate unintended
results.  Nevertheless, the use of
intelligence evidently leads to public value destroying consequences. 

 

The examples in this essay all portray the
destruction of public value as a result of performance measurement.  However, expected responses of performance
measurement advocates may argue that the unintended results illustrated in
these examples are preventable or manageable. 
Barber suggests that three things can be done to minimize unintended
consequences, first have a well-defined target, second anticipate as many
unintended effects as possible, and third regularly review the data collection
process to manage unintended consequences as they occur (Barber, 2015).  Moreover, Hood may also argue that public value
was destroyed in the mentioned examples because the selected measurement method
did not align with the culture of the organization.  If a more appropriate measurement method were
chosen, efficiency would not have been increased at the expense of public
value.  Although it is possible that unintended
consequences can be managed and anticipated, in practice, this is extremely
difficult and costly.  Given the large number
of people in an organization, the effects of performance measurement will vary,
making it extremely challenging to anticipate all consequences. It is also very
costly to identify and mitigate all the unintended consequences.  The organization may lack both the time and
financial resources to do so.  More
importantly, implementing a strategy to minimize the unintended consequences
may create unintended effects of its own.  In response to Hood’s claim that perverse
effects occurred because the measurement method chosen is not suitable to the
culture of the organization, often times identifying the right match between
method and culture is not possible.  Not
every organization can be categorized into hierarchical, fatalist, or
individualist, as some can be a mixture of more than one type.  Other external and internal aspects of the
organization are also important to consider when determining the measurement
method.  Therefore, determining
measurement method based on organizational culture may not be sufficient to
ensure unintended consequences will not occur. 

 

The
rise of performance management in the public sector has, in many ways, changed
the way public services are operated. 
Despite numerous claims of its performance-enhancing effects, it is evident
that efficiency is increased at the expense of public value.  The examples conveyed in this essay
illustrate that regardless of the measurement method used, the occurrence of
unintended consequences as a result of performance measurement leads to perverse
effects that destroy public value.  Performance
management provides a distorted incentive for members of the organization to
alter their behaviors in a way that is unfavorable.  The higher the impact performance measurement
has on the organization, the more prevalent unintended consequences are.  Given that measurement is not as
straightforward in the public sector as it is in the private sector, utilizing
public management techniques such as performance management increases
efficiency at the expense of creating public value.